发电技术 ›› 2019, Vol. 40 ›› Issue (2): 161-167.DOI: 10.12096/j.2096-4528.pgt.18173

• 火电及环境保护 • 上一篇    下一篇

汽轮机变工况下调节级压力预测模型及应用

罗云1,2(),陈雪林1(),李瑞东3(),苏永健1,*(),徐义巍1,晁俊凯1,李鹏竹1,任海彬1   

  1. 1 宁夏京能宁东发电有限责任公司, 宁夏回族自治区 银川市 750001
    2 北京东方国信科技股份有限公司, 北京市 朝阳区 100012
    3 北京京能电力股份有限公司, 北京市 朝阳区 100025
  • 收稿日期:2018-12-11 出版日期:2019-04-30 发布日期:2019-05-09
  • 通讯作者: 苏永健
  • 作者简介:罗云(1981),男,高级工程师,从事火力发电厂运行管理和大数据应用研究工作, ayunmeng@163.com|陈雪林(1984),男,工程师,从事火力发电厂运行和数据建模工作, 445752152@qq.com|李瑞东(1974),男,高级工程师,从事电力安全生产和技术管理工作, lird@sina.com

Prediction Model and Application of Turbine Regulating Stage Pressure Under Variable Conditions

Yun LUO1,2(),Xuelin CHEN1(),Ruidong LI3(),Yongjian SU1,*(),Yiwei XU1,Junkai CHAO1,Pengzhu LI1,Haibin REN1   

  1. 1 Ningxia Ningdong Power Generation Co., Ltd., Yinchuan 750001, Ningxia Hui Autonomous Region, China
    2 Beijing Jing Business-intelligence of Oriental Nations Corporation Ltd., Chaoyang District, Beijing 100012, China
    3 Beijing Jingneng Power Co., Ltd., Chaoyang District, Beijing 100025, China
  • Received:2018-12-11 Published:2019-04-30 Online:2019-05-09
  • Contact: Yongjian SU

摘要:

为监视汽轮机通流部分的健康状况和进行故障早期诊断,以汽轮机变工况计算为理论基础,基于弗留格尔公式和Weierstrass逼近定理,建立了调节级压力的多元回归模型。根据某660MW间接空冷机组大修后的运行数据对模型进行回归分析和验证,结果表明:模型具有很高的拟合度,自变量对因变量影响显著;稳态工况时,预测结果与实测值趋势一致,相对误差约为1.2%,实现了变工况下调节级压力软测量。将模型应用于电厂分布式控制系统中,建立了汽轮机通流部分故障的预警系统。

关键词: 汽轮机, 故障诊断, 调节级压力, 预测模型, 多元回归分析, 分布式控制系统(DCS)

Abstract:

Using the calculation of the variable conditions for steam turbines as the theoretical basis, a multivariate regression model for regulating pressure was established based on the Frugell formula and the Weierstrass approximation theorem. According to the historical operation data of a 660MW generating set, the model was subjected to regression analysis and verification. The results show that the model has a high degree of fit, and the independent variable has a significant influence on the dependent variable; When the steady state conditions are met, the predicted results are consistent with the measured values and the relative error is about 1.2%, which enable soft pressure measurement of the regulation stage. The model was applied to the distributed control system (DCS) of the power plant, and an on-line warning system was established to adjust the pressure anomaly and the partial flow malfunction in the turbine under variable conditions.

Key words: steam turbine, fault diagnosis, regulation stage pressure, predicted model, multiple regression analysis, distributed control system(DCS)